Petal
Smart Content Recommendations

Leverage Smart Content Recommendations to Build Relevancy with Your Customers

Take advantage of personalization powered by artificial intelligence (AI) with Kentico Kontent Smart Content Recommendations to connect your customers to the content they love.

Avatar fallback - no photo

Lukas TurekPublished on Feb 8, 2019

Everybody is talking about personalization these days, as it allows brands to serve relevant content to their customers. Personalization often features in Content as a Service (CaaS) platforms, not only helping you with content production and delivery but also enabling you to optimize content. 

Maybe you have also spent some time talking about providing personalized experiences to your customers, but never really implemented any advanced form of personalization. Based on discussions with Kentico Kontent customers and the options that are out there in the market, I can guess why.

We often hear that personalization is part of project plan, but it's usually skipped by other priorities simply because it takes a lot of development and testing time. And that can be true. When it’s done manually, you need to define segments of your audience, usually with many rules for each of those segments. Plus, you have to categorize your content based on who the target persona is.

As you can imagine this solution is complicated to maintain long term. As a consequence, we often then see the creation of segment definitions whose authors themselves do not know what it means after some time.

Better and Easier Personalization

We want to offer you a better and easier way to personalize your content so that you can build loyalty and relevancy with your customers. What if you could say “give me the next best content for this user given the interaction history” as a form of smart personalization that works completely autonomously? That’s exactly the power of the new recommendation engine in Kentico Kontent. You only need to select which content you want to recommend, tell that to the Recommendation API, and the underlying AI system analyzes all the content you have and provides you with the most relevant content for each of your users. And the best thing is, to personalize your content this way, all you need is just one API call compared to all the work described above.

Kentico Kontent Integrates Seamlessly with Recombee

For the recommendation itself, we joined forces with Recombee, specialists in artificial intelligence. We then built a Content Recommendation API on top of the recommendation engine as a layer that ties Kentico Kontent content with the service and takes care of data initialization, content updates and getting your recommendations.

Kontent.ai


We caught up with the founder and CEO of Recombee, Pavel Kordik, and wanted to share what he had to say about Recombee and Smart Content Recommendations:

Could you briefly introduce Recombee? 

Recombee is a leading hi-tech Recommender System as a Solution Provider. Recombee recommender engine provides high-quality personalization as a service for global clients from multiple domains. By using the user’s historical behavior, we can provide accurate suggestions that increase conversion rates by an average of 40% compared to the traditional list of best selling items. What is more, our sophisticated Artificial Intelligence self improves over time optimizing its performance based on preset KPIs. 

What is the real-time content recommendation and what results can our customers achieve?

Our solution analyzes user interactions and behavior online as well as product attributes and generates recommendations which are most likely to spark the interest of the customer. Our recommendations are updated in real-time to ensure the most accurate content for each visitor.

A real-time system is a system that processes input data within milliseconds so that the processed data is available almost immediately for feedback. So you can:

  • Create high-quality personalized recommendations
  • Deliver personalization within days, not months
  • Own the moment and offer real-time customer engagement
  • Increase Conversion Rate
  • Increase Basket Size 
  • Improve User Experience
  • Reduce Churn and Confusion

Do you have a success story you would like to highlight?

With the strategic goal to fully personalize its offer and help their shoppers to discover products they desire efficiently, Design Group felt it was necessary to ‘step up’ the personalization game and chose Recombee to deliver a complex recommendation solution tailored to their site’s https://moodings.com/ needs. For the +10.000 products, they needed to personalize moodings.com product offer while recognizing preferences of individual users, balance personalization and discovery of inspiring products and find similar products. A real-time response in large traffic was a must-have.

Solution

  • Automatic personalized recommendations applied to product view (Also See other Interesting Products sections) as well as in the shopping cart (Others Have Also Bought section).
  • Model for every single user with real-time updates.
  • Advance image processing models added to an ensemble of algorithms chosen for moodings.com 
  • Automated feature engineering.

Cooperation of Design Group and Recombee lead to:

  • 52% increase in the number of purchases since the implementation of Recombee.
  • Increasing relevancy of recommendations for individual store visitors.
  • Savings in the shop’s employees’ time previously spent on continuous manual product selection.
  • Savings in money spent on manual content selection.
  • Positive impact on shopper experience.

I can see that Recombee is developing very fast. What are your next plans?

We are currently conducting progressive research in image processing using a method of Deep Learning. Using artificial neural networks, we can recommend visually similar products based on a high-level understanding of the images. In some scenarios, our novel algorithm was able to outperform state-of-the-art methods of collaborative filtering in terms of CTR/CR. 

Where do you see the strongest potential in content recommendations for Kentico Kontent customers?

The strongest potential in content recommendations for Kentico Kontent customers lays in Recombee recommender engine's ability to enable smart personalization for content-heavy sites. Unlike traditional rule-based personalization systems, Recombee's AI-driven solution reflects real-time changes and complexity of user behavior online and thus improves user experience and increases your KPI's.

See It In Action 

Check out the Kentico Kontent blog where we recommend three similar articles under every blog post. One of our partners, Luminary, have also integrated Smart Content Recommendations to personalize content across their website (www.luminary.com). They actively recommend similar content in the blog, case studies, and employee sections. We'll be sharing more about their experience of using content recommendations in an upcoming blog post. 

Get Personalizing Your Content Today

We invite you to join the beta program so that you can start personalizing your content today. It's free to join, and you will get a chance to try Smart Content Recommendations before we launch it to all of our customers. To get started, contact us. We're more than happy to talk more about this new feature and answer any questions you may have.

Avatar fallback - no photo
Written by

Lukas Turek